Due to the unique anatomy of the posterior thoracic spine, providing instrumentation for applying corrections to the spine, through for example fusion devices, presents numerous challenges. Some instrumentation for engaging in various spinal corrective procedures includes nonsegmental hook constructs, such as Harrington rods. The implementation of these rods has potential adverse results, such as flat-back deformity, hook pull out or dislodgement and high nonunion rates. Other hook constructs, such as Cotrel-Dubosset, allow for segmental fixation and better curve correction and control; however, these constructs cause a relatively high pseudoarthrosis rate to occur due to the large number of hooks displacing bone grafts, which are required with these constructs for biological fusion to take place. Such systems have also proven to be exceptionally difficult to remove or revise. Hybrid constructs may be implemented that use pedicle screw instrumentation in the distal thoracic spine and hook fixation or sublaminar wire fixation in the proximal and mid-thoracic spine, but are still complicated by the aforementioned issues related to the hook and screw implementation described.
Thoracic pedicle screw instrumentation implementation has increased due to the increase in construct rigidity and scoliosis correction. Pedicle screw constructs may have a risk of neurological or vascular injury if they are misplaced. Studies have found that the more proximal screws were at greater risk of malposition where the pedicles may have abnormal morphology. Other studies have found that on average over 12% of screws were misplaced of which half of those were of concern (adjacent or impinging the aorta or other viscera, or within the spinal canal adjacent or impinging the spinal cord). To overcome the risks of thoracic pedicle screw instrumentation, some authorities suggest the use of intraoperative CT scans. Successful image guidance navigation systems may need to visualize the entire thoracic spine during deformity surgery. This typically requires 4 or 5 intraoperative CT scans for a typical patient. The relatively high radiation exposure of multiple CT scans, when used in various patients such as the typical female adolescent, is a concern as this is the time when these patients are at greatest risk for radiation induced breast or other cancers given that they are still growing and unshielded.
In elderly patients or adults with osteoporosis, pedicle screw constructs have a potential risk of screw pullout and may need augmentation with bone cement. Some spine surgeons may opt to use sublaminar wires or hook instrumentation at the end vertebra to reinforce the pedicle screws. However, even well placed screws are a risk for osteoporosis patients due to the potential of “plowing” of the pedicle screws out of the pedicles and subsequently affecting the adjacent vascular structures.
Another potential problem in deformity patients treated with pedicle screw constructs is an increased risk for proximal junctional kyphosis, PJK. There is a loss of thoracic kyphosis associated with thoracic pedicle screw constructs. Chronic PJK may be secondary to this iatrogenic thoracic lordosis as the spine tries to balance itself in patients treated with screw constructs.